Convolutional Neural Networks

Project: Write an Algorithm for a Dog Identification App


In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.

In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.


Why We're Here

In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

Sample Dog Output

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!

The Road Ahead

We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.

  • Step 0: Import Datasets
  • Step 1: Detect Humans
  • Step 2: Detect Dogs
  • Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
  • Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
  • Step 5: Write your Algorithm
  • Step 6: Test Your Algorithm

Step 0: Import Datasets

Make sure that you've downloaded the required human and dog datasets:

Note: if you are using the Udacity workspace, you DO NOT need to re-download these - they can be found in the /data folder as noted in the cell below.

  • Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dog_images.

  • Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.

Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.

In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.

In [3]:
import numpy as np
from glob import glob

# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))

# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images.
There are 8351 total dog images.

Step 1: Detect Humans

In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.

OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.

In [4]:
import cv2                
import matplotlib.pyplot as plt                        
%matplotlib inline                               

# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')

# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# find faces in image
faces = face_cascade.detectMultiScale(gray)

# print number of faces detected in the image
print('Number of faces detected:', len(faces))

# get bounding box for each detected face
for (x,y,w,h) in faces:
    # add bounding box to color image
    cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1

Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

Write a Human Face Detector

We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.

In [5]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray)
    return len(faces) > 0

(IMPLEMENTATION) Assess the Human Face Detector

Question 1: Use the code cell below to test the performance of the face_detector function.

  • What percentage of the first 100 images in human_files have a detected human face?
  • What percentage of the first 100 images in dog_files have a detected human face?

Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.

Answer: (You can print out your results and/or write your percentages in this cell)

In [6]:
from tqdm import tqdm

human_files_short = human_files[:100]
dog_files_short = dog_files[:100]

#-#-# Do NOT modify the code above this line. #-#-#

## TODO: Test the performance of the face_detector algorithm 
## on the images in human_files_short and dog_files_short.
count_hum_faces = 0
for hu_img in human_files_short:
    if face_detector(hu_img):
        count_hum_faces+=1

count_dog_faces = 0
for dog_img in dog_files_short:
    if face_detector(dog_img):
        count_dog_faces += 1
        
print("Human faces: %d percent" % count_hum_faces)
print("Doc faces: %d percent" % count_dog_faces)
Human faces: 98 percent
Doc faces: 17 percent

We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [7]:
### (Optional) 
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.

Step 2: Detect Dogs

In this section, we use a pre-trained model to detect dogs in images.

Obtain Pre-trained VGG-16 Model

The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.

In [8]:
import torch
import torchvision.models as models

# define VGG16 model
VGG16 = models.vgg16(pretrained=True)

# check if CUDA is available
use_cuda = torch.cuda.is_available()

# move model to GPU if CUDA is available
if use_cuda:
    VGG16 = VGG16.cuda()
    print("cuda used")
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /root/.torch/models/vgg16-397923af.pth
100%|██████████| 553433881/553433881 [00:08<00:00, 61815165.84it/s]
cuda used

Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.

(IMPLEMENTATION) Making Predictions with a Pre-trained Model

In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.

Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.

In [9]:
from matplotlib.pyplot import imshow
import numpy as np
from PIL import Image
import torchvision.transforms as transforms


def VGG16_predict(img_path):
    '''
    Use pre-trained VGG-16 model to obtain index corresponding to 
    predicted ImageNet class for image at specified path
    
    Args:
        img_path: path to an image
        
    Returns:
        Index corresponding to VGG-16 model's prediction
    '''
    
    ## TODO: Complete the function.
    ## Load and pre-process an image from the given img_path
    ## Return the *index* of the predicted class for that image
    img = Image.open(img_path)
    transform = transforms.Compose([
        transforms.Resize(256),
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])
    img_tensor = transform(img)
    input = img_tensor.unsqueeze(0)
    
    if use_cuda:
        input = input.cuda()
    
    output = VGG16(input)
    probs = torch.nn.functional.softmax(output, dim=1)
    _, idx = torch.max(probs, dim=1)
    return idx.item() # predicted class index

VGG16_predict(dog_files[0])
Out[9]:
243

(IMPLEMENTATION) Write a Dog Detector

While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).

Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).

In [10]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
    ## TODO: Complete the function.
    idx = VGG16_predict(img_path)
    return idx >= 151 and idx <= 268 # true/false

(IMPLEMENTATION) Assess the Dog Detector

Question 2: Use the code cell below to test the performance of your dog_detector function.

  • What percentage of the images in human_files_short have a detected dog?
  • What percentage of the images in dog_files_short have a detected dog?

Answer:

In [11]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
hum_as_dog = 0

for human_img in human_files_short:
    if dog_detector(human_img):
        hum_as_dog += 1

hum_percent = hum_as_dog * 100 / len(human_files_short)
print("human as dog detected %d" %hum_percent)

dog_as_dog = 0

for dog_img in dog_files_short:
    if dog_detector(dog_img):
        dog_as_dog += 1

dog_percent = dog_as_dog * 100 / len(dog_files_short)
print("dog as dog detected %d" %dog_percent)
    
human as dog detected 0
dog as dog detected 100

We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [40]:
### (Optional) 
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.

Step 3: Create a CNN to Classify Dog Breeds (from Scratch)

Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.

We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.

Brittany Welsh Springer Spaniel

It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).

Curly-Coated Retriever American Water Spaniel

Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.

Yellow Labrador Chocolate Labrador Black Labrador

We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.

Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!

In [12]:
import os
import numpy as np
from torchvision import datasets
from glob import glob
from PIL import Image, ImageFile
from torch.utils.data import Dataset
import torchvision.transforms as transforms
from torch.utils.data import DataLoader
import matplotlib.pyplot as plt
import cv2

ImageFile.LOAD_TRUNCATED_IMAGES = True

transform_train = transforms.Compose([
    transforms.Resize(300),
    transforms.CenterCrop(300),
    transforms.RandomRotation(20),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])

transform_valid = transforms.Compose([
    transforms.Resize(300),
    transforms.CenterCrop(300),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])

transform_test = transforms.Compose([
    transforms.Resize(300),
    transforms.CenterCrop(300),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])

dataset_valid = datasets.ImageFolder('/data/dog_images/valid', transform_valid)
dataset_train = datasets.ImageFolder('/data/dog_images/train', transform_train)
dataset_test = datasets.ImageFolder('/data/dog_images/test', transform_test)

#
# Did not know what in review is meant by: Implement three different DataLoader
#

dl_valid = DataLoader(dataset_valid, batch_size=20, shuffle=True)
dl_test = DataLoader(dataset_test, batch_size=20, shuffle=True)
dl_train = DataLoader(dataset_train, batch_size=20, shuffle=True)

loaders_scratch = {
    'valid': dl_valid,
    'test': dl_test,
    'train': dl_train
}

Question 3: Describe your chosen procedure for preprocessing the data.

  • How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
  • Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?

Answer:

  • resize and cropping by 300x300 because it seems suitable, that the NN is not so big and dont need to many layers mit MaxPooling.
  • I choose a randomRotation for prevent overfitting. It seems that the validation_loss decrease is more stable with randomRotation

(IMPLEMENTATION) Model Architecture

Create a CNN to classify dog breed. Use the template in the code cell below.

In [14]:
import torch.nn as nn
import torch.nn.functional as F
import torch


# check if CUDA is available
use_cuda = torch.cuda.is_available()

# define the CNN architecture
class Net(nn.Module):
    ### TODO: choose an architecture, and complete the class
    def __init__(self):
        super(Net, self).__init__()
        ## Define layers of a CNN
        self.c1 = nn.Conv2d(3, 16,3)
        self.c2 = nn.Conv2d(16,32,3)
        self.c3 = nn.Conv2d(32,64,3)

        self.l1 = nn.Linear(64*35*35, 600)
        self.m1 = nn.MaxPool2d(2,2)
        self.do = nn.Dropout(0.25)
        self.l2 = nn.Linear(600, 134)
        self.batch_norm = nn.BatchNorm1d(num_features=600)
    
    def forward(self, x):
        ## Define forward behavior
        x = self.m1(F.relu(self.c1(x))) #600x600
        x = self.do(x)
        x = self.m1(F.relu(self.c2(x))) #300x300
        x = self.m1(F.relu(self.c3(x))) #150x150

        x = x.view(-1,64*35*35) #16x16
        x = self.batch_norm(self.l1(x))
        x = F.relu(x)
        x = self.l2(x)
        return x

#-#-# You so NOT have to modify the code below this line. #-#-#

# instantiate the CNN
model_scratch = Net()

# move tensors to GPU if CUDA is available
if use_cuda:
    model_scratch.cuda()

Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.

Answer: I built up a neural network with three conv2d layer incresing features from 3 to 64 following by 2 linear layers. I tried out a different number of conv layers up to 5. At the end 3 layers fits the requirement to get a accuracy > 10%. Between the conv2d layer the NN downsize the images with Maxpool2d. The downsizing where done to get small first linear layer. From starting 3 x 300 x 300 = 270.000 to 643535 = 78.400 inputs. More inputs without downsampling results in out of memory The batch_norm of l1 helps makes the NN more stable and the valid_loss decreases more faster. The dropout after first convulational layer helps to prevent overfitting

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.

In [15]:
import torch.optim as optim

### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()

### TODO: select optimizer
optimizer_scratch = torch.optim.SGD(model_scratch.parameters(), lr=0.03)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.

In [53]:
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    valid_loss_min = np.Inf 
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0

        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## find the loss and update the model parameters accordingly
            ## record the average training loss, using something like
            ## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            optimizer.zero_grad()

            output = model(data)

            loss = criterion(output, target)
            train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            loss.backward()

            optimizer.step()
            #print(train_loss)
            
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)

            loss = criterion(output, target)
            valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        if valid_loss_min > valid_loss:
            valid_loss_min = valid_loss
            torch.save(model.state_dict(), save_path)
            print("Saved model ")
    # return trained model
    return model


use_cuda = torch.cuda.is_available()
# train the model
model_scratch = train(15, loaders_scratch, model_scratch, optimizer_scratch, 
                      criterion_scratch, use_cuda, 'model_scratch.pt')

# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
Epoch: 1 	Training Loss: 4.626222 	Validation Loss: 4.519699
Saved model 
Epoch: 2 	Training Loss: 4.220038 	Validation Loss: 4.485821
Saved model 
Epoch: 3 	Training Loss: 3.963184 	Validation Loss: 4.201240
Saved model 
Epoch: 4 	Training Loss: 3.705399 	Validation Loss: 4.162113
Saved model 
Epoch: 5 	Training Loss: 3.435168 	Validation Loss: 3.951767
Saved model 
Epoch: 6 	Training Loss: 3.118216 	Validation Loss: 3.851619
Saved model 
Epoch: 7 	Training Loss: 2.785616 	Validation Loss: 3.813555
Saved model 
Epoch: 8 	Training Loss: 2.402189 	Validation Loss: 3.942541
Epoch: 9 	Training Loss: 2.012308 	Validation Loss: 3.859097
Epoch: 10 	Training Loss: 1.650437 	Validation Loss: 3.907293
Epoch: 11 	Training Loss: 1.281885 	Validation Loss: 4.070237
Epoch: 12 	Training Loss: 0.969640 	Validation Loss: 4.032326
Epoch: 13 	Training Loss: 0.726567 	Validation Loss: 3.963401
Epoch: 14 	Training Loss: 0.556642 	Validation Loss: 4.192157
Epoch: 15 	Training Loss: 0.439947 	Validation Loss: 4.131852

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.

In [16]:
def test(loaders, model, criterion, use_cuda):

    # monitor test loss and accuracy
    test_loss = 0.
    correct = 0.
    total = 0.

    model.eval()
    for batch_idx, (data, target) in enumerate(loaders['test']):
        # move to GPU
        if use_cuda:
            data, target = data.cuda(), target.cuda()
        # forward pass: compute predicted outputs by passing inputs to the model
        output = model(data)
        # calculate the loss
        loss = criterion(output, target)
        # update average test loss 
        test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
        # convert output probabilities to predicted class
        pred = output.data.max(1, keepdim=True)[1]
        # compare predictions to true label
        correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
        total += data.size(0)
            
    print('Test Loss: {:.6f}\n'.format(test_loss))

    print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
        100. * correct / total, correct, total))

    
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
# call test function    
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.819426


Test Accuracy: 13% (110/836)

Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)

You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).

If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.

In [17]:
## TODO: Specify data loaders

    
transform_trans_train = transforms.Compose([
    transforms.Resize(224),
    transforms.CenterCrop(224),
    transforms.RandomRotation(20),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])

transform_trans_valid = transforms.Compose([
    transforms.Resize(224),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])


transform_trans_test = transforms.Compose([
    transforms.Resize(224),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
    ])




dataset_valid_trans = datasets.ImageFolder('/data/dog_images/valid', transform_trans_valid)
dataset_train_trans = datasets.ImageFolder('/data/dog_images/train', transform_trans_train)
dataset_test_trans = datasets.ImageFolder('/data/dog_images/test', transform_trans_test)

dataloader_valid = DataLoader(dataset_valid_trans, batch_size=20, shuffle=True)
dataloader_train = DataLoader(dataset_train_trans, batch_size=20, shuffle=True)
dataloader_test = DataLoader(dataset_test_trans, batch_size=20, shuffle=True)

loaders_transfer = {
    'valid': dataloader_valid,
    'test': dataloader_test,
    'train': dataloader_train,
}

(IMPLEMENTATION) Model Architecture

Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.

In [18]:
import torchvision.models as models
import torch.nn as nn

## TODO: Specify model architecture 
model_transfer = models.vgg16(pretrained=True)

for param in model_transfer.features.parameters():
    param.requires_grad = False
    
model_transfer.classifier[6] =  nn.Linear(4096, 134)

print(model_transfer)

if use_cuda:
    model_transfer = model_transfer.cuda()
VGG(
  (features): Sequential(
    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU(inplace)
    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (3): ReLU(inplace)
    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (6): ReLU(inplace)
    (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (8): ReLU(inplace)
    (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (11): ReLU(inplace)
    (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (13): ReLU(inplace)
    (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (15): ReLU(inplace)
    (16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (18): ReLU(inplace)
    (19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (20): ReLU(inplace)
    (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (22): ReLU(inplace)
    (23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (25): ReLU(inplace)
    (26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (27): ReLU(inplace)
    (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (29): ReLU(inplace)
    (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (classifier): Sequential(
    (0): Linear(in_features=25088, out_features=4096, bias=True)
    (1): ReLU(inplace)
    (2): Dropout(p=0.5)
    (3): Linear(in_features=4096, out_features=4096, bias=True)
    (4): ReLU(inplace)
    (5): Dropout(p=0.5)
    (6): Linear(in_features=4096, out_features=134, bias=True)
  )
)

Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.

Answer:

Only removed the last linear layer and replaced it with a layer where the output fits to the number of solutions. The VGG16 NN is trained over the ImageNet picture lib and able to detect much more types of pictures that is needed in this solution. So it should also be able to easliy detect the dog breed. The Conv Layers are not touched and with requires_grad = False we prevent a retraining. So only the linear layers are trained for this solution. The pictures have to transformed to size of 224x224.

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.

In [19]:
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.SGD(model_transfer.classifier.parameters(), lr=0.001)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.

In [ ]:
use_cuda = torch.cuda.is_available()
# train the model
model_transfer = train(15, loaders_transfer, model_transfer, optimizer_transfer, 
                      criterion_transfer, use_cuda, 'model_transfer.pt')
Epoch: 1 	Training Loss: 3.982301 	Validation Loss: 2.481360
Saved model 
Epoch: 2 	Training Loss: 2.013998 	Validation Loss: 1.116722
Saved model 
Epoch: 3 	Training Loss: 1.278164 	Validation Loss: 0.762691
Saved model 
Epoch: 4 	Training Loss: 0.985004 	Validation Loss: 0.628125
Saved model 
Epoch: 5 	Training Loss: 0.857255 	Validation Loss: 0.561026
Saved model 
Epoch: 6 	Training Loss: 0.764692 	Validation Loss: 0.537614
Saved model 
Epoch: 7 	Training Loss: 0.698872 	Validation Loss: 0.502152
Saved model 
Epoch: 8 	Training Loss: 0.646531 	Validation Loss: 0.477804
Saved model 
Epoch: 9 	Training Loss: 0.593853 	Validation Loss: 0.454229
Saved model 
Epoch: 10 	Training Loss: 0.572601 	Validation Loss: 0.444468
Saved model 
Epoch: 11 	Training Loss: 0.532931 	Validation Loss: 0.450234
Epoch: 12 	Training Loss: 0.508134 	Validation Loss: 0.420728
Saved model 
Epoch: 13 	Training Loss: 0.488338 	Validation Loss: 0.433217
Epoch: 14 	Training Loss: 0.465491 	Validation Loss: 0.407639
Saved model 
In [20]:
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.

In [21]:
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.458024


Test Accuracy: 86% (719/836)

(IMPLEMENTATION) Predict Dog Breed with the Model

Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.

In [22]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
from PIL import Image

# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in dataset_train_trans.classes]

def predict_breed_transfer(img_path):
    # load the image and return the predicted breed
    img = Image.open(img_path).convert('RGB')
    transform = transforms.Compose([
        transforms.Resize(224),
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
        ])

    data = transform(img)[:3,:,:].unsqueeze(0)
    data = data.cuda()
    result = model_transfer(data)

    _, idx = torch.max(result, dim=1)

    return class_names[idx]

predict_breed_transfer("/data/dog_images/train/059.Doberman_pinscher/Doberman_pinscher_04159.jpg")
Out[22]:
'Doberman pinscher'

Step 5: Write your Algorithm

Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,

  • if a dog is detected in the image, return the predicted breed.
  • if a human is detected in the image, return the resembling dog breed.
  • if neither is detected in the image, provide output that indicates an error.

You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.

Some sample output for our algorithm is provided below, but feel free to design your own user experience!

Sample Human Output

(IMPLEMENTATION) Write your Algorithm

In [29]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.

def run_app(img_path):
    ## handle cases for a human face, dog, and neither
    if dog_detector(img_path):
        return "dog", predict_breed_transfer(img_path)
    elif face_detector(img_path):
        return "human", predict_breed_transfer(img_path)
    else:
        return "no dog and no human detected", None

Step 6: Test Your Algorithm

In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?

(IMPLEMENTATION) Test Your Algorithm on Sample Images!

Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.

Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.

Answer: It's better than expected.

  1. more epochs to get more accurcity
  2. use more random transforms to prevent overfitting
  3. clean up code ;-)
In [32]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
from PIL import Image
from IPython.display import display # to display images

## suggested code, below
for file in np.hstack((human_files[:3], dog_files[:3])):
    pil_im = Image.open(file)
    display(pil_im)
    t, result = run_app(file)
    print("This is a", t, "recognized as", result)
    print("--------------------------------------")
This is a human recognized as Brittany
--------------------------------------
This is a human recognized as Dachshund
--------------------------------------
This is a human recognized as Afghan hound
--------------------------------------
This is a dog recognized as Bullmastiff
--------------------------------------
This is a dog recognized as Mastiff
--------------------------------------
This is a dog recognized as Mastiff
--------------------------------------
In [34]:
#testimages from Unsplash
test_files = np.array(glob("myTestImages/*"))

for file in test_files:
    pil_im = Image.open(file)
    display(pil_im)
    t, result = run_app(file)
    print("This is a", t, "recognized as", result)
    print("--------------------------------------")
This is a human recognized as Chesapeake bay retriever
--------------------------------------
This is a human recognized as Dogue de bordeaux
--------------------------------------
This is a dog recognized as Bernese mountain dog
--------------------------------------
This is a human recognized as Chinese crested
--------------------------------------
This is a dog recognized as Dandie dinmont terrier
--------------------------------------
This is a dog recognized as Collie
--------------------------------------
In [ ]: